mbox series

[0/1] HID: Add introduction about HID for non-kernel programmers

Message ID d6d16821-2592-8210-475a-5388d7a79e82@polimi.it (mailing list archive)
Headers show
Series HID: Add introduction about HID for non-kernel programmers | expand

Message

Marco Morandini June 20, 2023, 12:32 p.m. UTC
This doc addition is meant to allow a random user 
to understand what is going wrong with his 
brand-new device, and possibly try to fix it on his own.
I've written it, being one of those random users,
in an effort to document what I think I've learned,
and with the hope to minimize the pain to guys like me.

The chapter is by no means complete, and for sure will require
careful checking and fixing from the HID maintainers.
In my opinion, it would also be great if they could add a brief
explanation for the different quirks (I've left a FIXME there).
I hope I've not misunderstood too many concepts (?events?) and 
that the whole thing is not so screwed up that it's better to
throw it into the thrash can and run away without looking back.

Marco

Comments

Benjamin Tissoires June 20, 2023, 12:59 p.m. UTC | #1
Hi Marco,

On Jun 20 2023, Marco Morandini wrote:
> 
> This doc addition is meant to allow a random user 
> to understand what is going wrong with his 
> brand-new device, and possibly try to fix it on his own.
> I've written it, being one of those random users,
> in an effort to document what I think I've learned,
> and with the hope to minimize the pain to guys like me.

And many thanks for that. This is well written IMO and addresses a lot
of the pain points. So thanks!

> 
> The chapter is by no means complete, and for sure will require
> careful checking and fixing from the HID maintainers.
> In my opinion, it would also be great if they could add a brief
> explanation for the different quirks (I've left a FIXME there).

For that particular part (I'll reiterate in the review of the patch),
I'm not sure we are tackling the problem at the correct place. We should
document those in the code directly, and include/reference them in the
.rst file. Trying to duplicate this information is prone to
desynchronisation of the code and the doc and probably noone will keep
it up to date when there is a change.

> I hope I've not misunderstood too many concepts (?events?) and 
> that the whole thing is not so screwed up that it's better to
> throw it into the thrash can and run away without looking back.

There are a few nitpicks here and there, but you got the whole thing
(almost) right IMO :)

Cheers,
Benjamin